Assessing Gaze and Pointing: Human Cue Interpretation by Indian Free-Ranging Dogs in a Food Retrieval Task

TL;DR

This study demonstrates that Indian free-ranging dogs can interpret combined human pointing and gazing cues to find hidden food, emphasizing the importance of cue redundancy in interspecific communication.

Contribution

It provides novel evidence that free-ranging dogs rely on multimodal human cues, highlighting the role of signal clarity and dog demeanor in cue comprehension.

Findings

Dogs succeeded only with combined cues, not unimodal or conflicting.

Affiliative dogs showed higher success rates and shorter approach latencies.

Demeanor influenced engagement but not cognitive ability to interpret cues.

Abstract

The urban habitat provides a landscape that increases the chances of human-animal interactions, which can lead to increased human-animal conflict, but also coexistence. Some species show high levels of socio-cognitive abilities that enable them to perceive communicational gestures of humans and use them for their own benefit. This study investigated the ability of Indian free-ranging dogs (Canis lupus familiaris) to utilise human social-referential cues (pointing and gazing) to locate hidden food, focusing on the relative effectiveness of unimodal versus multimodal cues. A total of 352 adult free-ranging dogs were tested in an object-choice task involving six different cue conditions: control (no cue), negative control (one baited bowl, no cue), combined pointing and gazing, pointing-only, gazing-only, and conflicting cues (pointing and gazing at opposite bowls). The dogs successfully chose the correct target only in the combined pointing and gazing condition, while performance under unimodal and conflicting cue conditions did not differ significantly from chance. This highlights the importance of signal redundancy and clarity in interspecific communication for this population. A dog's demeanor was a significant predictor of its willingness to engage: affiliative dogs were significantly more likely to succeed in the overall experiment and displayed a significantly shorter approach latency compared to anxious and neutral dogs. While demeanor affected the approach latency, it did not affect the accuracy of the choice, decoupling the dogs' personality from its cognitive ability to comprehend the clear cue. Neither the dogs' sex nor the experimental condition significantly predicted approach latency.